# Are labels informative in semi-supervised learning? -- Estimating and leveraging the missing-data mechanism

arXiv: https://arxiv.org/abs/2302.07540 · HTML: https://mlsanity.com/paper/2302.07540/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 49 references that are in the map.
- First posted on arXiv 15 Feb 2023.

## Builds on (most cited first, 12 of 12)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1902.03368/) (2019)
- [ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring](https://mlsanity.com/paper/1911.09785/) (2019)
- [In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning](https://mlsanity.com/paper/2101.06329/) (2021)
- [CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning](https://mlsanity.com/paper/2102.09559/) (2021)
- [Open-World Semi-Supervised Learning](https://mlsanity.com/paper/2102.03526/) (2021)
- [Linear predictor on linearly-generated data with missing values: non consistency and solutions](https://mlsanity.com/paper/2002.00658/) (2020)
- [On Non-Random Missing Labels in Semi-Supervised Learning](https://mlsanity.com/paper/2206.14923/) (2022)
- [Kernel Machines With Missing Responses](https://mlsanity.com/paper/1806.02865/) (2018)
- [On missing label patterns in semi-supervised learning](https://mlsanity.com/paper/1904.02883/) (2019)

## Built on by (most cited first, 5 of 5)

- [Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets](https://mlsanity.com/paper/2401.14497/) (2024)
- [From Biased Selective Labels to Pseudo-Labels: An Expectation-Maximization Framework for Learning from Biased Decisions](https://mlsanity.com/paper/2406.18865/) (2024)
- [Model Evaluation in the Dark: Robust Classifier Metrics with Missing Labels](https://mlsanity.com/paper/2504.18385/) (2025)
- [SSLfmm: An R Package for Semi-Supervised Learning with Mixed Missingness](https://mlsanity.com/paper/2512.03322/) (2025)
- [Semi-Supervised Classification with Informative Missing Labels in Weibull Mixture Models](https://mlsanity.com/paper/2609.00774/) (2026)

## Abstract (arXiv)

Semi-supervised learning is a powerful technique for leveraging unlabeled data to improve machine learning models, but it can be affected by the presence of ``informative'' labels, which occur when some classes are more likely to be labeled than others. In the missing data literature, such labels are called missing not at random. In this paper, we propose a novel approach to address this issue by estimating the missing-data mechanism and using inverse propensity weighting to debias any SSL algorithm, including those using data augmentation. We also propose a likelihood ratio test to assess whether or not labels are indeed informative. Finally, we demonstrate the performance of the proposed methods on different datasets, in particular on two medical datasets for which we design pseudo-realistic missing data scenarios.